Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 84, 1956-57
Previous Article | Table of Contents | Next Article
This text is also available in PDF
(2 MB) Opens in new window
– 321 –

Sex and Condition of Fish

The “condition factor” or weight-length relationship of snapper will be discussed in detail in a further paper (Cassie, in press). Although there is a distinct seasonal fluctuation in the condition factor, the variation between individual fish is so

– 322 –

great that the changes cannot be readily detected except by statistical means.* No correlation has been detected between the sexual condition of the gonad (as classified in Table 4) and the condition factor of the fish. Although no specific investigation of palatability was carried out, fish taken during the investigation were regularly eaten aboard the ship. It was found that snapper which had apparently finished spawning were seldom in any way inferior eating to those with full, ripe gonads. Both these observations seem to conflict with the belief that snapper, immediately after spawning, are normally in a thin condition, with soft flesh easily damaged by handling, and often unattractive or unpalatable. The term “spent” is commonly applied to fish in such condition, apparently in the belief that depletion of the gonads is concurrent with physical deterioration, perhaps in analogy with the Pacific salmon which dies very shortly after spawning. It is, of course, obvious that the discharge of eggs or milt would result in some loss of weight (seldom more than 5 per cent, of the total weight of the fish), but on the other hand this loss may either be replaced almost immediately by additional food, or be negligible compared with variations due to other causes. Thus, although the “spent” fish is no doubt a reality to the fish merchant and may occur more commonly after the spawning season, it is none the less clear that physical deterioration and loss of weight do not inevitably follow after spawning.

In order to record the condition of the gonads in samples dissected, the code shown in Table 4 was employed.

The degree of certainty with which the different classes can be determined is variable. Condition IV could always be identified with little margin for doubt, as could Condition III in the case of females. Ripe eggs are readily seen through the membrane of the ovary, being transparent and about twice the size of the opaque

Table 4.—Gonad Condition Code.
Code Brief Description. Female. Male.
I Immature Ovary small and immature (usually a young fish). Testes small and immature (usually a young fish).
II Mature Ovary firm and well developed but with no ripe eggs. Testes well developed but milt not discharged by pressure on abdomen of fish.
III Ripe Ripe eggs present but cannot be discharged by pressure on abdomen of fish.
IV Running Eggs readily discharged by gentle pressure on abdomen of fish. Milt readily discharged by gentle pressure on abdomen of fish.
V Spent Ovary small and watery in appearance. Testes small and watery in appearance.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table 5.—Monthly Distribution of Sexual Condition
Year. Month I II III IV V N I II & III IV V N
1950 November 9 76 13 2 45 71 29 21
December 6 44 16 6 27 81 7 72 18 3 61
1951 January 7 7 87 45 24 76 17
October 21 30 49 100 12 88 140
November 5 18 65 10 1 17 4 87 10 82
December† – 17 61 22 18 15 77 8 13

[Footnote] * For any one sample of snapper the relationship between length, L, and weight, W, may be described by the equation: W = KLb the exponent, b, has a value of approximately 2.7, while the constant, K, may be used as a measure of condition factor in place of the usual factor, C, where: C = W/L3

[Footnote] † Taken by handline. N = Total number of fish examined.

– 323 –

yellow immature eggs. Conditions I and V were sometimes indistinguishable, and in cases of doubt were classed as V. The results are shown in Table 5. Catches were, for the most part, taken in the western portion of the Gulf, north of Tiritiri Island, but other catches taken from various other regions have been included in the monthly average since they do not appear to differ in their characteristics to any marked degree. Although the data are not sufficiently comprehensive to give a detailed picture of change in sexual conditions, there appear to be several features worthy of mention. As might be expected, there is a preponderance of mature or ripe gonads at the beginning, and of spent gonads toward the end of the season. There is, however, a relatively small number of running fish, and this group is never the dominant one in the catch. There are three possible reasons for this. First, that the time during which eggs or milt may be discharged is of relatively short duration (although it is noticeable that the proportion of running males is usually greater than that of.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table 6.—Sex Ratio.
Date. Males. Females χ2 P
29 11.50 21 27 0.750 0.3 - 0.5
5.12.50 28 25 0 170 0.5 - 0.7
12.12.50 3 4 0 143 0.7 - 0.8
14 12.50 18 25 1.140 0.2 - 0.3
15.12.50 12 27 5.769 0.01 - 0.02
4. 1.51 3 20 12.565 < 0.001
5. 1.51 14 25 3.103 0.05 - 0.1
12. 1.51 39 41 0.050 0.8 - 0.9
17 1.51 7 26 10 939 < 0.001
19. 1.51 9 7 0 250 0.5 - 0.7
31. 1.51 16 21 0.676 0.3 - 0.5
13 13 0.000 0.9
5. 3.51 7 13 1.800 0.1 - 0.2
14. 3.51 15 5 5.000 0.02 - 0.05
15 3.51 13 13 0.000 > 0.9
1. 851 18 10 2.286 0.1 - 0.2
29. 8.51 22 18 0.400 0.5 - 0.7
6. 9.51 19 14 0.758 0.3 - 0.5
27. 9.51 21 18 0.231 0.5 - 0.7
(b) 23.10.51 22 18 0.400 0.5 - 0.7
24.10.51 24 16 1.600 0.2 - 0.3
25.10.51 18 22 0.400 0.5 - 0.7
25 10.51 23 17 0.900 0.3 - 0.5
30.10 51 27 13 4.900 0.02 - 0.05
31.10 51 26 14 3.600 0.05 - 0.1
1.11.51 20 20 0.000 > 0.9
1.11 51 18 22 0.400 0.5 - 0.7
14 11.51 20 19 0.026 0.8 - 0.9
28 11.51 24 16 1.600 0.2 - 0.3
18 12 51 8 10 0.222 0.5 - 0.7
18. 8.52 18 20 0.105 0.7 - 0.8
4.9 52 16 21 0.676 0.3 - 0.5
23.10 52 21 19 0.100 0.7 - 0.8
30.11.52 15 25 2.500 0.1 - 0.2
Total (d.f. = 34) 63.459 < 0.001
Pooled (d.f. = 1) 598 624 0 553 0.3 - 0.5

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Interaction (d.f. = 33) 62 906 < 0.001
d.f. χ2 P
Total 10 34.805 < 0.001
(a) Pooled 1 14.011 < 0.001
Interaction 9 20 794 < 0.02
Total 10 15 475 0.1 - 0.2
(b) Pooled 1 9.473 0.002
Interaction 9 6.002 0.7 - 0.8
– 324 –

females). Secondly, that the spawning snapper may tend to rise into midwater (out of range of the trawl) for the act of spawning but remain near the bottom at other times. Thirdly, that rough treatment received in the trawl may cause the ripe sexual products to be discharged before landing so that running fish are no longer recognisable when examined. Some support is given to the last two possibilities by the fact that in the December, 1952 sample (the only one taken by handlines), 77 per cent of the males and 22 per cent, of the females are running. About half of these fish were taken in midwater, and the line usually subjects the fish to less rough handling than the trawl.

Comparison between the two seasons must be treated with caution, since the only two sets of data taken under strictly comparable conditions are those for November. Even making due allowance for the obvious deficiencies of the data, it is difficult to reconcile Table 5 with the frequency of eggs as shown in Fig. 10. Thus, although spawning was earlier in 1950–51, the condition of female fish as judged by columns II and III seems to be less advanced in the months of November and December. It is also surprising that spent fish do not predominate until January, 1951, two months after the principal burst of spawning. Although a more comprehensive collection of data might shed more light on these apparent anomalies, it is likely that neither trawl nor handline will give a truly representative sample of the entire spawning population if fish in the different stages have a different behaviour pattern—e.g., migration, vertical distribution, susceptibility to capture, etc.

From the fish samples described above, together with data taken at other times of the year, estimates of the male-female ratio were made. Table 6 gives this data together with a χ2 test which gives a measure of departure from the expected ratio of unity.

For the pooled data the overall departure from expectation is not significant, but both the total and interaction χ2 are significant at the 0.1 per cent. level, indicating heterogeneity in the sex ratio between samples. The high level of significance is contributed principally by the two samples, 4/1/51 and 17/1/51. If these are removed from the table, neither total nor interaction χ2 is significant, suggesting that the two samples are in some way aberrant. However, there is further evidence of heterogeneity in the two periods (a) and (b). In the first of these (December, 1950, and January, 1951) there is a consistently greater number of females, while in the second (August to October, 1951) males predominate. The separate analysis at the bottom of the table shows that in both cases the ratio of the pooled sex totals deviates significantly from unity (χ2 = 14.011, 9.473), but the periods differ in that (a) shows heterogeneity in the sex ratio (χ2 = 20 794) while (b) does not (χ2 = 6.002). Although no permanent disparity between male and female snapper is evident from Table 6, there is apparently either a differential mortality or (far more probably) a difference in behaviour which results in a greater number of one or other other sex being caught in certain seasons.

Previous Article | Table of Contents | Next Article